Chen Ma 1, Min Li 1, Yufei Shi 2, Xiao Zhu 1
1 Fudan University (Shanghai, China), 2 The Chinese University of Hong Kong (Hong Kong, China)
Objectives
Bacteriophage therapy in immunocompetent hosts is governed not only by phage replication and bacterial dynamics, but also by host immune responses. While most PKPD models incorporate immune-mediated bacterial killing, immune-driven attenuation of phage is rarely represented explicitly. This omission may distort exposure–response interpretation under repeated dosing.
To develop a mechanism-based PKPD framework that quantitatively characterizes host immunity as a bidirectional determinant of therapeutic outcome-simultaneously enhancing bacterial clearance while limiting functional phage availability-and to assess the resulting nonlinear exposure-response behavior arising from opposing immune effects.
Methods
A nonlinear mixed-effects PKPD model was constructed using a coupled system of ordinary differential equations describing bacterial subpopulations, phage pharmacokinetics, and immune effects. Model development integrated in vitro phage-bacteria kinetic data and preclinical in vivo infection data to inform parameter estimation. Model evaluation included parameter precision assessment and visual predictive checks (VPCs).
Phage pharmacokinetics followed a two-compartment structure with lytic production from infected bacteria.Host-associated reduction in effective phage availability following administration was represented by a time-dependent input fraction:
F(t) = exp(-kmac × time)
where kmac quantifies immune-associated attenuation of functional phage input.
Bacterial dynamics were described using a structured population model consisting of susceptible (S), early infected (I₁), late infected (I₂), and resistant (R) subpopulations, with total bacterial burden defined as:
N = S + I1 + I2 + R
Susceptible bacteria followed logistic growth with density dependence and were subject to phage adsorption and immune-mediated elimination:
dS/dt = α×S×(1 – N/Bmax) – β×S×Cp – Kill×S
Infection progression was represented through transit compartments, with lytic production characterized by burst size and lysis rate parameters informed by in vitro experiments.
Immune-mediated bacterial elimination was modeled as a capacity-limited process with delayed activation:
Kill = k × (N/(KI+N)) × (1/(1+exp(-kimm×(t-t50))))
Results
The model adequately described repeated-dose phage pharmacokinetics and bacterial burden dynamics. Incorporation of time-dependent phage input attenuation was necessary to reproduce the progressive decline in systemic phage exposure observed experimentally. Sensitivity analyses indicated that immune-associated reduction in phage availability influenced systemic exposure, whereas immune-mediated bacterial killing contributed to bacterial suppression.
Exploratory simulations suggested that immune effects act in opposing directions within the system, affecting both phage availability and bacterial dynamics.
Conclusion
This mechanism-based PKPD framework demonstrates that host immunity exerts bidirectional effects in phage therapy by simultaneously influencing effective phage availability and bacterial elimination. Explicit representation of these mechanisms improves structural consistency of exposure-response interpretation. Future work will evaluate structural identifiability and explore integration of immune effects on phage and bacteria within a unified immune-driven component to better reflect physiological coherence and system-level regulation.
References:
[1] Raphaëlle Delattre et al., Cell Report. 2022 May 17;39(7):110825.
[2] Gauri G Rao et al., Clin Pharmacol Ther. 2025 Jan;117(1):94-105.
Reference: PAGE 34 (2026) Abstr 12285 [www.page-meeting.org/?abstract=12285]
Poster: Drug/Disease Modelling - Infection